NEW YORK (GenomeWeb) – Whole-genome sequencing of healthy individuals and cardiomyopathy patients by and large does not appear to increase follow-up medical care and costs in the short term, according to preliminary results from Brigham and Women's Hospital's MedSeq project presented at the American Society of Human Genetics annual meeting in Baltimore last week.
In addition, most study participants seem to be increasingly comfortable contributing their genetic results to their electronic medical records as the study progresses, the researchers found.
The goal of the ongoing study, funded in 2012 under the National Human Genome Research Institute's Clinical Sequencing Exploratory Research (CSER) program, is to explore the integration of genome sequencing into the practice of medicine.
Under the study's protocol, the researchers have recruited a total of 200 participants, 100 cardiomyopathy patients and 100 apparently healthy individuals between the ages of 40 and 65. Half the patients in each group were randomized to obtain and discuss a report of their family medical history with a primary care physician, the other half to discuss the same report plus an interpretation of their genome sequence, focusing on medically relevant variants in 4,600 genes that are associated with Mendelian disease, recessive carrier status, pharmacogenomics, and polygenic risk for eight cardiometabolic traits.
This summer, MedSeq investigators reported early findings on how primary care physicians communicated genome sequencing results to patients, finding that doctors did not make egregious errors in reporting the information, despite not being experts in genetics.
Three talks at ASHG by MedSq investigators last week focused on the short-term effects of genomic results on medical care and associated costs, and how comfortable participants are with genomic information going into their electronic health records.
Many hope that sequencing the genomes of healthy individuals might help in the prevention and early detection of disease. On the other hand, genome results may lead to unnecessary follow-up tests and doctor visits that do not make a difference to a person's health and might even be harmful. According to Jason Vassy, one of the project's co-principal investigators, in order to determine the risks and benefits of genome sequencing, it will be important to measure its short-term and long-term effects.
For their analysis, he and his colleagues measured the short-term outcomes of genome sequencing, focusing on the healthy patient cohort of the MedSeq study. They looked at how genome results affected decisions by primary care providers, whether follow-up care was in line with prevention guidelines from the US Preventive Services Task Force, and what types of healthcare services participants had received six months after obtaining their reports, including their cost.
A week after disclosing results to patients, primary care physicians were surveyed about the follow-up medical care they had ordered as a result of the report. Six months later, the researchers examined patients' electronic health records to see what tests, follow-up visits, or other medical care they had received and whether it complied with USPSTF guidelines. They also determined the cost of these services, using billing data and Medicare price weights.
So far, 64 of the 100 healthy participants have reached the six-month follow-up mark. Overall, most patients who obtained genome information did not get additional medical care compared to those who did not, Vassy reported. In particular, the two groups did not differ significantly in the number of laboratory tests, imaging tests, medications, and specialist visits their primary physicians ordered.
Genome recipients did, however, have more orders for cardiac tests, most often electrocardiograms, than family-history-only recipients, likely in response to a polygenic risk of heart disease that was suggested by the genome results, he said. The study did not provide polygenic risks for other conditions, such as cancer or neurological disease, he noted, which might have prompted additional interventions by physicians.
A chart review six months later showed that the two groups did not differ in how their healthcare followed preventive care guidelines, such as colorectal cancer screening and aspirin use, or in the number of imaging tests or specialist visits.
Genome sequencing also did not add to participants' short-term healthcare costs. The mean cost was $1,161 per genome recipient, compared to $825 per family-history-only participant, which Vassy said was not statistically different. Median costs were $257 per genome recipient, compared to $239 for non-genome recipients.
An economic analysis, presented at ASHG by MedSeq investigator Kurt Christensen, looked at the short-term costs of genome sequencing, compared to family history only, in both the healthy and the cardiovascular disease cohorts of the MedSeq study. The analysis included 108 participants, about two thirds from the healthy cohort and a third from the cardiovascular patient group.
The study included costs for consenting, genome sequencing, Sanger confirmation, genome interpretation and reporting, and the disclosure meeting, as well as follow-up healthcare costs over a six-month period as determined from the EHR. The time required to review variants and write the genome report has dropped as the process has become more efficient, Christensen noted, making it less expensive. Disclosure sessions, on average, took 15 minutes longer for participants with genome data compared to those without.
Mean total costs amounted to approximately $9,500 for genome recipients, compared to about $2,300 for recipients of family history results only. Predictably, the lion's share of this — almost $6,000 — was related to genome sequencing.
On average, genome recipients had about $3,600 in follow-up medical care costs, compared to $2,200 for non-genome recipients, a difference Christensen said was not statistically significant. Median costs for medical care were $740 for genome recipients versus $360 for non-genome recipients, and cardiology patients, overall, had higher medical costs than primary care participants.
Future analyses will seek to incorporate healthcare costs outside of the Partners HealthCare System that participants are part of, and to include other measures besides healthcare costs, for example, time off from work. Also, depending on funding, it might be possible to determine the impact of genome sequencing over the long term. Whether or not long-term benefits from sequencing exist that might justify its cost is "an open question," Christensen said.
Finally, Carrie Blout, a genetic counselor and project manager for the MedSeq project, presented results on the attitudes of MedSeq participants about the incorporation of whole-genome sequencing results in their electronic medical records, and how they changed over time.
All MedSeq participants — who are mostly highly educated, middle class, and non-Hispanic white — were consented to have both their family history report and their genome results placed in their EMR. Of the more than 500 individuals the researchers initially recruited for the study, about a third declined to participate in the end, Blout noted, citing fear of insurance discrimination, psychosocial issues related to the results, general privacy concerns, and other reasons.
MedSeq participants were surveyed twice about their attitude and comfort level with having genome data in the EMRs, once before they learned whether they would obtain genomic results and another time six weeks after they received their report.
Initially, over half of participants said they agreed that genome results should be part of the EMR. Six weeks after disclosure, that number had grown to 66 percent.
Also, after six weeks, 77 percent of those participants who received their genome results said they were "very comfortable" contributing these to their EMR, an increase from a little over half at baseline.
Finally, 65 percent of genome recipients were willing to have all their genetic information placed in their EMR after six weeks, compared to 45 percent before they received their results.
Overall, participants' comfort with placing genome results in their EMR appeared to increase over the course of the study, Blout said, and there was not a difference in preference between the healthy and the cardiology cohorts.